The Amorphous Cube (aka ‘morpho’) was built for me by its inventor, Dario DiBattista. It retains a roughly cube-like shape but eliminates colors to identify the sides. In this respect it’s a little like an uncolored bump cube.

For a description of how the cube was built see this thread in the Puzzle Building forum.




All parts are irregularly shaped, and every single cubie is unique. Some faces are rounded, some are slightly concave. It has a very solid feel, yet it’s lightweight. It’s finished in a glossy blue lacquer with a pebbly texture.



Even when the cube is scrambled, the exposed sides of the cubies are finished in the same blue lacquer as the outer faces. This gives it a refined appearance whether solved or scrambled.



The turning surfaces has been carefully sanded free of snags, and the cube turns with gentle, almost silky resistance. There’s just the right amount of resistance that it holds its shape at any time.

Despite its origin, conventional 3x3 solutions obviously don’t work. Pieces must be solved by matching groups of adjacent cubies. Corners must be solved after the adjacent edges, and the edges can only be solved by matching them to correctly oriented centers. It’s harder than it sounds because most 3x3 sequences mess up the center orientations, so it’s necessary to devise new sequences for just about every type of move.

The Amorphous Cube resembles the familiar 3x3 so it was comfortable to play with, yet almost every piece required new techniques to solve. It was familiar but entirely new.

Very clever mod. I'd love to get my hands on one of these.



Why not? As long as you can solve the supercube, the knowledge can easily be applied here using blockbuilding alongside some BLD-esque piece movers.

Then, once you have the 'feel' for where each piece goes, surly you can use a conventional supercubed-up method to solve it?

Very nice. I would love to try one of these. All the extra putty doesn't add too much weight then?

That's what I thought at first.

But a supercube isn't a conventional cube (that's the point of any supercube) and requires more techniques than a conventional cube.

Some solvers approach a supercube using conventional methods, ignoring center orientations until the last. That's impractical for this puzzle because you wouldn't be able to reliably use the centers to verify cubie placement. There are enough "decoy" matches, where cubies incorrectly appear to be matched that solving quickly stalls. It would be somewhat more practical to manually correct the center orientations whenever they need it, but it makes solving tediously long.

So you're right, solving this cube rquires a thorough mastery of supercube methods and princples, no shortcuts. And it warrants some strategy changes: Block building works overall, but the cubies are most efficiently solved in a different order within each block.

And of course, almost every sequence temporarily separates the pieces that have been solved, making them difficult to find if you lose your orientation. It has to be handled to be believed!

Bondo putty contains very lightweight fillers, so it weighs something between a Pyraminx and an old-style Revenge:

99 g = Rainbow Cube (7 color)
120 g = Generic 3x3x3
144 g = Pyraminx
173 g = Amorphous Cube
176 g = Meffert's Megaminx (tiled)
202 g = Rubik's Revenge (1982 Macau model)
252 g = Meffert's Master Cube (tiled)
305 g = Rubik's Wahn
311 g = Rubik's 5x5x5 (current)
328 g = Meffert's Professor Cube (tiled)

Normally I don't find shape changing puzzles attractive, but with this one... I want it.

I've never seen one like it. It has an almost organic look to it. Thank you for sharing.

Anyone think we'll see other puzzles done like this?

That is a simple and cool build

Here's a picture to illustrate why this thing requires ingenuity to solve.

In the picture below, the entire top face is solved except the nearest corner. Now, how would you go about figuring out which corner cubie goes here?

Answer: it is the correct one already, it's just the wrong orientation. But that's difficult to detect visually. It could've just as easily been one of the other four corners, except I set it up this way for the photo. Some of the "wrong" corner cubies would've looked just like this.

So it's a bigger challenge to identify the correct pieces than to move them.



I sure hope so, which is why I posted the construction details in the Puzzle Building section. Each one is likely to be a unique, and probably have a different skill level depending on how distinctive the shapes turn out. A 2x2 would be more difficult to build and probably harder to solve, too.

I can understand it being more difficult to solve, but how would it be harder to make one?

This would be so awesome made from a square-1. If someone does do that, please post a picture.

The puzzle has to be disassembled & reassembled at various stages of construction.

With a Rubik's 2x2 it would be hard to disassemble/reassemble using the twist method because the parts couldn't flex as much, and they'd probably be hard to grip. It might be necessary to maintain access to the assembly screw, an extra step.

An Eastsheen 2x2 would be even harder to safely disassemble after the putty was applied, so it would probably be necessary to disassemble it ahead of time and hold the parts on a custom-made jig to apply the putty.

Beautiful! Reminds me of a pebble you would find on a beach.

If only I could find one of these lying on a beach :)

Nice job!

The funny thing is, if it operates like a normal 3x3, then no moves are ever blocked. This is funny because of the shape changing, it looks like it would almost block moves due to the odd shapes of the pieces.

Genius design, although I would probably just get frustrated with it.

That's true.


That's true.


That's true!!

Here's a picture showing another one of the challenges. The top face has been mostly solved, but A and B don't match even though they seem to fit their other neighbors. Which cubie is wrong one, A or B? Could B and C both be wrong? What now??

Most of the cubies are very distinct, but there are just enough confusing situations like this to keep it interesting!

The cool thing about that puzzle is that no matter how many times you build one, none of them will be the same. Its like the snowflake of the puzzle world.

Not unless you started mass producing them, in that case you COULD make them all the same. But what fun would that be?

I would like to see one in a green marble color, green owns all!

I know I've already posted in this thread, but I keep coming back to it. It is definitely my #1 of the moment. To me it looks simple and complex at the same time, and just so smooth!

my current fav., too! i want one BAD!
although, unlike others have said, i find it very ugly, but in a charming, strange way. a gross, ugly puzzle.
i keep staring, and i understand that not having colors might throw you off from what goes where, but it seems like, especially if you use the beginners method, it shouldnt as hard as verywetpaint says.
he would know better than me, and i believe him about the difficulty, but still, doesnt seem that hard just by looking at it/thinking about it.
i REALLY wanna find out for myself...

Look at it this way: When you scramble the cube, the pieces aren't in order any more. Some pieces will just have a slightly different concave, and look like they SHOULD work, even though they don't.

LBL would probably be the way to go with this puzzle, Petrus would be a bit hard for it, especially orienting bad edges

Petrus is probably your best bet, working in blocks would benefit it more if you ask me.


@VWP: can I request a photo of the checkerboard pattern please?

Great idea. I've attached a photo that concentrates on one face in the checkerboard pattern. Yes, I've positively verified that it's the standard checkerboard pattern using three half-turns of the center slices, equivalent to six face half-turns.

I originally expected to see all the edge cubies meet each other at the same heights, and all the corner cubies to meet the center at the same height. The picture shows the unexpected result.

Why don't the corners meet the center at the same height?

They're from the same face, but the checkerboard is equivalent to turning each face 180 degrees, so each center gets turned around. The red dots mark where the lower-right corner and center would meet on the solved cube.

Why don't the edge pieces meet at the same height?

In making the checkerboard pattern, each edge piece has been turned around as it traveled to the opposite face. The green dots mark where the lower and left edge cubies should meet on a solved cube.